Legal claims defining the scope of protection. Each claim is shown in both the original legal language and a plain English translation.
1. A payment document classification system, comprising: a communication unit to receive an image of a payment document captured by a camera of a mobile device; a feature knowledge database to store a set of features, wherein a given feature in the set of stored features is associated with one or more probability values, and each of the one or more probability values represents a probability of occurrence of the given feature in a given payment document type within a set of known payment document types; a preprocessing unit to extract a set of features from the image; a classification unit which is configured to: compare, by one or more processors, the set of extracted features against the set of stored features in the feature knowledge database; in response to a match between an extracted feature in the set of extracted features and a stored feature in the set of stored features, determine whether the payment document can be classified as one of the known payment document types associated with the matched stored feature based on the associated probability values of the matched stored feature by comparing the probability values associated with the matched stored feature with a predetermined threshold value; and classify the payment document as the payment document type based on the determination: a processing unit which is configured to: process content of the payment document in view of the payment document type.
This system classifies payment documents using image analysis and probabilistic feature matching. The technology addresses the challenge of accurately identifying different types of payment documents (e.g., invoices, receipts, checks) from images captured by mobile devices, which is critical for automated financial processing. The system includes a communication unit that receives an image of a payment document from a mobile device camera. A feature knowledge database stores predefined features, each linked to probability values indicating the likelihood of that feature appearing in specific payment document types. A preprocessing unit extracts features from the input image. A classification unit compares these extracted features against the stored features. If a match is found, the system evaluates the associated probability values against a threshold to determine if the document can be classified as one of the known types. If the threshold is met, the document is classified accordingly. A processing unit then handles the document content based on its identified type, enabling specialized processing for different document formats. This approach improves accuracy in document classification by leveraging probabilistic feature matching rather than rigid rule-based systems.
2. The payment document classification system of claim 1 , wherein the classification unit is configured to determine whether the payment document can be classified as one of the known payment document types by: in response to one of the probability values associated with the matched stored feature exceeds the predetermined threshold value, determining that the payment document can be classified as the known payment document type associated with the one of the probability values.
This technical summary describes a payment document classification system designed to automatically categorize payment documents into predefined types. The system addresses the challenge of efficiently and accurately classifying diverse payment documents, such as invoices, receipts, or payment confirmations, to streamline financial processing. The classification system includes a feature extraction unit that analyzes the payment document to identify key features, such as text patterns, numerical values, or structural elements. These extracted features are compared against a database of stored features associated with known payment document types. Each stored feature is linked to a probability value indicating the likelihood that a document containing that feature belongs to a specific type. The classification unit evaluates the probability values of the matched stored features. If any of these probability values exceeds a predetermined threshold, the system determines that the payment document can be classified as the corresponding known payment document type. This threshold-based approach ensures that only highly probable matches are accepted, reducing misclassification errors. The system may also include a training unit that updates the stored features and probability values based on new data, improving classification accuracy over time. By automating the classification process, the system enhances efficiency in financial workflows, reducing manual review and processing time.
3. The payment document classification system of claim 1 , wherein the set of stored features includes one or more of a geometric feature, a lock icon, a Magnetic Ink Character Recognition (MICR) line, an OCR-A line, a check number, a barcode, a keyword, and a combination of two or more of the above.
The payment document classification system is designed to automatically categorize different types of payment documents, such as checks, invoices, and receipts, by analyzing their visual and textual features. The system addresses the challenge of manually sorting and processing diverse payment documents, which is time-consuming and prone to errors. To solve this, the system extracts and stores a set of distinctive features from each document, which are then used to classify the document type accurately. The stored features include geometric characteristics, such as the document's shape and layout, as well as specific visual elements like lock icons, Magnetic Ink Character Recognition (MICR) lines, OCR-A lines, check numbers, barcodes, and keywords. These features are selected because they are commonly found in payment documents and can reliably distinguish between different types. For example, a MICR line is typically present on checks but not on invoices, while a barcode may appear on receipts. The system can also analyze combinations of these features to improve classification accuracy. By leveraging these stored features, the system automates the classification process, reducing manual effort and increasing efficiency in financial document processing.
4. The payment document classification system of claim 1 , wherein the set of known payment document types includes at least personal checks, business checks, cashier's checks, traveler's checks, money orders, store rebates, gift certificates, and IRS refunds.
The payment document classification system is designed to automatically identify and categorize different types of payment documents, addressing the challenge of manually sorting diverse financial instruments in banking, retail, and financial services. The system processes digital images or scanned copies of payment documents and classifies them into predefined categories. These categories include personal checks, business checks, cashier's checks, traveler's checks, money orders, store rebates, gift certificates, and IRS refunds. The system uses machine learning or pattern recognition techniques to analyze visual and textual features of the documents, such as logos, formats, and printed text, to determine the document type accurately. This automation reduces processing time, minimizes human error, and improves efficiency in financial transactions. The system may also integrate with existing document management or payment processing workflows to streamline operations further. By supporting a wide range of payment document types, the system ensures broad applicability across different industries and use cases.
5. The payment document classification system of claim 1 , wherein the classification unit compares the set of extracted features against the set of stored features in the feature knowledge database in a sequence until one of the set of extracted features matches one of the set of stored features.
The payment document classification system is designed to automatically categorize different types of payment documents, such as invoices, receipts, and payment confirmations, by analyzing their features. The system addresses the challenge of manually sorting and classifying large volumes of payment documents, which is time-consuming and prone to errors. To solve this, the system extracts features from a payment document, such as text patterns, numerical values, and layout structures, and compares these against a pre-existing database of stored features associated with known document types. The classification unit processes the extracted features in a sequential manner, checking each feature against the stored database until a match is found. Once a match is identified, the document is classified accordingly. This approach ensures accurate and efficient categorization, reducing manual effort and improving processing speed. The system may also include preprocessing steps to enhance feature extraction, such as optical character recognition (OCR) for scanned documents, and may use machine learning techniques to refine the feature database over time. The sequential comparison method ensures that even partially matching documents are correctly classified, improving reliability in real-world applications.
6. The payment document classification system of claim 4 , wherein the classification unit compares the set of extracted features against the set of stored features in the feature knowledge database in a series, starting with an extracted feature which requires the least computation time among the set of extracted features.
A payment document classification system processes and categorizes payment documents by extracting features from the documents and comparing them against stored features in a feature knowledge database. The system includes a feature extraction unit that identifies and extracts relevant features from payment documents, such as text, numerical values, or structural elements. These extracted features are then passed to a classification unit, which compares them against a set of stored features in the database. The comparison process is optimized by prioritizing features that require the least computational time, ensuring efficient classification. The system may also include a feature knowledge database that stores predefined features and their associated document categories, allowing the classification unit to match extracted features to known categories. This approach improves accuracy and speed in classifying payment documents, such as invoices, receipts, or payment confirmations, by leveraging computational efficiency in feature comparison. The system may further include a preprocessing unit to prepare documents for feature extraction, such as noise reduction or format normalization, enhancing the reliability of the classification process.
7. The payment document classification system of claim 1 , further comprising a configuration unit to allow a user to configure the set of payment document types into a category of payment document types for classifying the payment document as belonging to the category of payment document types.
A payment document classification system categorizes and organizes payment-related documents, such as invoices, receipts, and statements, to streamline financial processing. The system uses machine learning or rule-based methods to analyze document features, such as text, layout, and metadata, to determine the type of payment document. This automation reduces manual effort and improves accuracy in document handling. The system includes a configuration unit that allows users to customize the classification process. Users can define and group payment document types into broader categories, enabling flexible categorization based on organizational needs. For example, a user may group "invoice" and "receipt" under a "vendor payment" category or "statement" and "remittance" under a "customer payment" category. This adaptability ensures the system aligns with specific business workflows and reporting requirements. The configuration unit may also support rule-based adjustments, such as setting thresholds for document similarity or defining custom classification criteria. By enabling user-defined categorization, the system enhances scalability and precision in document management.
8. The payment document classification system of claim 1 , further comprising an image correction unit to receive the image of the payment document from the communication unit and corrects at least one aspect of the image of the payment document to produce a corrected image.
A payment document classification system processes images of payment documents, such as invoices or receipts, to classify and extract relevant information. The system includes an image correction unit that receives the document image and enhances its quality by correcting distortions, improving clarity, or adjusting brightness/contrast. This correction ensures the document is in an optimal state for subsequent processing, such as optical character recognition (OCR) or machine learning-based classification. The corrected image is then used to accurately identify document types, extract key data fields, and organize financial records. The system addresses challenges in handling low-quality or improperly captured document images, which can lead to misclassification or data extraction errors. By improving image quality before analysis, the system enhances the reliability and efficiency of automated payment document processing in financial and accounting workflows.
9. A method of classifying a payment document, comprising: receiving an image of a payment document captured by a camera of a mobile device; extracting a set of features from the image; comparing the set of extracted features against a set of features stored in a feature knowledge database, wherein a given feature in the set of stored features is associated with one or more probability values, and each of the one or more probability values represents a probability of occurrence of the given feature in a given payment document type within a set of known payment document types; and in response to a match between an extracted feature in the set of extracted features and a stored feature in the set of stored features, determining whether the payment document can be classified as one of the known payment document types associated with the matched stored feature based on the associated probability values of the matched stored feature by comparing the probability values associated with the matched stored feature with a predetermined threshold value; classifying the payment document as the payment document type based on the determination; and processing content of the payment document in view of the payment document type.
The invention relates to a method for classifying payment documents using image analysis and feature matching. The problem addressed is the need for accurate and automated classification of various types of payment documents, such as invoices, receipts, or bills, captured by mobile devices. The method involves receiving an image of a payment document taken by a mobile device camera. Features are extracted from the image, such as text patterns, layout structures, or graphical elements. These extracted features are compared against a database of pre-stored features, where each stored feature is linked to one or more probability values. Each probability value indicates the likelihood of a given feature appearing in a specific type of payment document within a predefined set of known document types. When a match is found between an extracted feature and a stored feature, the system evaluates the associated probability values against a predetermined threshold to determine if the document can be classified as one of the known types. If the threshold is met, the document is classified accordingly. The classified document is then processed based on its type, enabling further actions like data extraction or payment initiation. This method improves the efficiency and accuracy of payment document handling in automated systems.
10. The method of claim 9 , wherein determining whether the payment document can be classified as one of the known payment document types includes: in response to one of the probability values associated with the matched stored feature exceeds the predetermined threshold value, determining that the payment document can be classified as the known payment document type associated with the one of the probability values.
This invention relates to automated classification of payment documents, such as invoices, receipts, or payment confirmations, to streamline financial processing. The challenge addressed is accurately identifying and categorizing diverse payment document formats, which often vary in structure, layout, and content, making manual or rule-based classification inefficient and error-prone. The method involves analyzing a payment document to extract features, such as text patterns, numerical values, or layout structures, and comparing these features against a database of known payment document types. Each known type is associated with a set of predefined features and corresponding probability values indicating the likelihood of a match. The system calculates probability values for each known type based on the similarity between the extracted features and the stored features. If any of these probability values exceeds a predetermined threshold, the payment document is classified as the corresponding known type. This probabilistic approach improves accuracy by reducing false classifications and handling variations in document formats. The method may also include preprocessing steps, such as noise reduction or feature normalization, to enhance matching accuracy. By automating classification, the system reduces manual effort, speeds up financial workflows, and minimizes errors in document processing.
11. The method of claim 9 , wherein the set of stored features includes one or more of a geometric feature, a lock icon, a MICR line, an OCR-A line, a check number, a barcode, a keyword, and a combination of two or more of the above.
The invention relates to document processing, specifically methods for analyzing and extracting information from documents such as checks. The problem addressed is the need for accurate and efficient identification of key features in documents to facilitate automated processing, such as check verification or data extraction. Traditional methods often struggle with varying document formats, low-quality scans, or obscured features, leading to errors in processing. The method involves storing a set of predefined features that are commonly found in documents like checks. These features include geometric patterns, security elements like lock icons, MICR (Magnetic Ink Character Recognition) lines, OCR-A (Optical Character Recognition) lines, check numbers, barcodes, and keywords. The system can also recognize combinations of these features to improve accuracy. By detecting and analyzing these features, the method enables automated systems to verify document authenticity, extract relevant data, and streamline processing workflows. The approach enhances reliability in document handling, reducing manual intervention and improving efficiency in applications like banking, financial services, and document management.
12. The method of claim 9 , wherein the set of known payment document types includes at least personal checks, business checks, cashier's checks, traveler's checks, money orders, store rebates, gift certificates, and IRS refunds.
This invention relates to automated payment document processing systems, specifically improving the accuracy and efficiency of identifying and categorizing different types of payment documents. The problem addressed is the difficulty in distinguishing between various payment document types, such as personal checks, business checks, cashier's checks, traveler's checks, money orders, store rebates, gift certificates, and IRS refunds, which often share similar visual features but require different processing workflows. The method involves analyzing a payment document to determine its type by comparing it against a predefined set of known document types. The system extracts visual and textual features from the document, such as layout, text patterns, logos, and security elements, and matches them against stored templates or classification models. For example, a personal check may be identified by its specific layout, payee line, and MICR (Magnetic Ink Character Recognition) line, while a money order may be distinguished by its unique serial number and issuer branding. The system then categorizes the document into one of the predefined types, enabling appropriate routing and processing. This approach ensures accurate classification, reducing manual intervention and errors in financial transactions. The method supports a wide range of payment documents, enhancing flexibility in automated document processing systems.
13. The method of claim 9 , wherein comparing the set of extracted features against the set of features includes performing the comparisons sequentially until one of the set of extracted features matches one of the set of stored features.
This invention relates to a method for comparing extracted features against a set of stored features in a sequential manner. The method is designed to improve the efficiency and accuracy of feature matching, particularly in applications such as image recognition, pattern matching, or data retrieval. The problem addressed is the need for a systematic and reliable way to compare features without unnecessary computational overhead, ensuring that a match is found as quickly as possible. The method involves extracting a set of features from input data, such as an image or a dataset. These extracted features are then compared sequentially against a predefined set of stored features. The comparison process continues until a match is found between one of the extracted features and one of the stored features. This sequential approach ensures that the comparison stops as soon as a match is detected, optimizing computational resources and reducing processing time. The method may be part of a larger system that includes preprocessing steps, such as feature extraction and storage, as well as post-processing steps, such as validation or further analysis. The sequential comparison ensures that the system efficiently identifies relevant matches without exhaustive searches, making it suitable for real-time applications where speed and accuracy are critical. The invention enhances the performance of feature-based matching systems by minimizing unnecessary comparisons and improving overall efficiency.
14. The method of claim 9 , wherein comparing the set of extracted features against the set of features includes performing the comparisons in a series, starting with an extracted feature which requires the least computation time among the set of extracted features.
This invention relates to a method for efficiently comparing extracted features against a set of features in a computational system. The problem addressed is the computational inefficiency in feature comparison processes, particularly in systems where multiple features must be evaluated against a reference set, such as in pattern recognition, image processing, or machine learning applications. The method involves extracting a set of features from input data, such as an image or sensor readings, and then comparing these extracted features against a predefined set of features stored in a database or reference model. The key innovation is in the order of comparisons: the method prioritizes comparisons starting with the extracted feature that requires the least computational time to evaluate. This approach minimizes overall processing time by deferring more computationally intensive comparisons until after simpler, faster comparisons have been completed. The method may also include preprocessing steps to determine the computational complexity of each extracted feature before comparison begins. This allows the system to dynamically adjust the comparison sequence based on real-time computational demands. Additionally, the method may incorporate early termination criteria, where the comparison process stops if a match is found before all features are evaluated, further optimizing efficiency. This technique is particularly useful in real-time systems where processing speed is critical, such as in autonomous vehicles, medical imaging, or industrial automation, where rapid and accurate feature matching is essential.
15. The method of claim 9 , wherein the method further comprises configuring the set of payment document types into a category of payment document types for classifying the payment document as belonging to the category of payment document types.
This invention relates to automated payment document classification in financial systems. The problem addressed is the inefficiency and errors in manually categorizing diverse payment documents, such as invoices, receipts, and remittance slips, which are critical for accounting and financial processing. The solution involves a method for classifying payment documents by analyzing their content and structure to determine their type, such as distinguishing between an invoice and a receipt. The method includes extracting features from a payment document, such as text patterns, numerical values, and layout structures, and comparing these features against predefined criteria to identify the document type. Additionally, the method configures payment document types into broader categories, allowing for hierarchical classification. For example, an invoice may be classified under a "financial transaction" category, while a receipt may belong to a "purchase confirmation" category. This categorization improves accuracy in financial workflows by ensuring documents are routed to the correct processing systems. The system may use machine learning models trained on labeled payment documents to enhance classification accuracy. The method also supports dynamic updates to classification rules based on new document types or evolving financial regulations. By automating this process, the invention reduces manual effort, minimizes errors, and accelerates financial operations.
16. The method of claim 9 , wherein the method further comprises correcting at least one aspect of the image of the payment document to produce a corrected image.
A system and method for processing payment documents, such as checks or invoices, involves capturing an image of the document and analyzing its content to extract relevant financial data. The method includes detecting and correcting distortions or errors in the captured image to improve accuracy. Image correction may involve adjusting brightness, contrast, or alignment, or removing noise or artifacts. The corrected image is then processed to extract data such as account numbers, amounts, or payee information. This ensures reliable automated processing of payment documents, reducing manual intervention and errors. The method may also include validating the extracted data against predefined rules or formats to further enhance accuracy. The system may be integrated into banking or financial software to streamline payment processing workflows.
Unknown
September 24, 2019
Browse 5M+ US patents with plain-English claim translations and AI-generated analysis.